AMD Radeon 535 Mobile vs NVIDIA GeForce GTX 850A
Processeur graphique
| Nom du GPU | Polaris 24 | GM107 |
|---|---|---|
| GPU variant | Polaris 24 XT (216-0867030) | N15P-GT-A1 |
| Architecture | GCN 3.0 | Maxwell |
| Fonderie | TSMC | TSMC |
| Taille du processus | 28 nm | 28 nm |
| Transistors | 1,550 million | 1,870 million |
| Taille des matrices | 125 mm² | 148 mm² |
Graphiques mobiles
| Date de publication | Apr 18th, 2017 | Mar 17th, 2014 |
|---|---|---|
| Génération | Mobility Radeon (M500) | GeForce 800A |
| Production | End-of-life | End-of-life |
| Bus interface | PCIe 3.0 x8 | PCIe 3.0 x16 |
| Prédécesseur | Crystal System | GeForce 700A |
| Successeur | — | GeForce 900A |
Vitesse clock
| Base clock | 730 MHz | 902 MHz |
|---|---|---|
| Boost clock | 1024 MHz | 936 MHz |
| Mémoire clock | 1125 MHz 4.5 Gbps effective | 900 MHz 1800 Mbps effective |
Mémoire
| Taille de la mémoire | 2 GB | 2 GB |
|---|---|---|
| Type de mémoire | GDDR5 | DDR3 |
| Bus mémoire | 64 bit | 128 bit |
| Bande passante | 36.00 GB/s | 28.80 GB/s |
Render Config
| Shading units | 384 | 640 |
|---|---|---|
| TMUs | 24 | 40 |
| ROPs | 8 | 16 |
| Compute units | 6 | — |
| Cache L1 | 16 KB (per CU) | 64 KB (per SMM) |
| Cache L2 | 128 KB | 2 MB |
| Comptage SMM | — | 5 |
Performance théorique
| Taux de pixel | 8.192 GPixel/s | 14.98 GPixel/s |
|---|---|---|
| Taux de texture | 24.58 GTexel/s | 37.44 GTexel/s |
| FP16 (half) performance | 786.4 GFLOPS (1:1) | — |
| FP32 (float) performance | 786.4 GFLOPS | 1,198 GFLOPS |
| FP64 (double) performance | 49.15 GFLOPS (1:16) | 37.44 GFLOPS (1:32) |
Board Design
| Taille Slot | IGP | IGP |
|---|---|---|
| TDP | 50 W | 45 W |
| Sorties | No outputs | No outputs |
| Connecteurs de puissance | None | — |
Caractéristiques graphiques
| DirectX | 12 (12_0) | 12 (11_0) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.0 | 3.0 |
| Vulkan | 1.2 | 1.1 |
| Shader model | 6.3 | 5.1 |
| CUDA | — | 5.0 |
